To find your market rate: pull data from three sources (Glassdoor, LinkedIn Salary, and one industry salary survey), adjust for your location and company size, then set your anchor at the 75th percentile for your level. This gives you a defensible number backed by current data rather than guesswork or what a colleague earns.
Most professionals walk into a salary negotiation with a number they picked up from a friend, a vague memory of a job advert, or worse, what they currently earn. None of those is your market rate. Your market rate is what an employer in your sector, at your level, in your location, would pay a qualified candidate today. The gap between that number and the one in your head is often larger than you think, and it always matters.
Market rate data is imperfect everywhere, but three sources consistently outperform the rest for UK, US, and Australian professionals.
Glassdoor. The largest crowdsourced salary database. Filter tightly: job title, location, company size, and whether the figure is base salary or total compensation. The numbers skew slightly high because self-reporting is voluntary, but the percentile spread is useful. Look at the 25th, 50th, and 75th percentile rather than the headline average.
LinkedIn Salary. Often overlooked, but the data is verified against real profiles, which makes it more reliable than anonymous self-reporting. The tool shows salary ranges by title, location, years of experience, and education level. Access it through the LinkedIn Salary tab. It requires a premium account in some regions, but the free tier shows enough to cross-reference.
Reed or Totaljobs (UK) / Seek (Australia) / Indeed (all three). Live job adverts are underused as market data. When an employer posts a salary band, that is real-time willingness to pay. Search your target role and filter by location. Record the bands from 10 to 15 adverts. This gives you current demand-side data, not historical survey responses.
Use all three. Do not rely on a single source.
Collect figures from all three sources and record them in a simple table: source, low end, midpoint, high end. You are looking for the midpoint of the credible range, not the maximum figure you can find.
Calculate a weighted midpoint. Give live job adverts the most weight (they reflect current demand), then LinkedIn (verified profiles), then Glassdoor (useful for spread but slightly inflated). A simple approach: average the midpoints across all three sources. That number becomes your baseline.
The baseline is not yet your anchor. You apply four adjustment factors before you settle on a figure.
Company size. Large employers (FTSE 100, Fortune 500, ASX 50) typically pay 10 to 20 per cent above market midpoint for equivalent roles. Startups and SMEs often pay below midpoint on base but compensate with equity or benefits. Adjust your anchor up or down accordingly.
Sector. Financial services, technology, and energy pay above the cross-sector median. Non-profit, education, and public sector typically pay below. If the role you are targeting is in a higher-paying sector than the average in your dataset, add a sector premium of 5 to 15 per cent.
Experience relative to the band. Market data represents the full range of people doing that job. If you are at the top end of the experience range for this role, your anchor should sit at the 65th to 75th percentile of the range, not the 50th. If you are newer to the level, stay closer to the midpoint.
Total compensation. Base salary is rarely the whole story. Pension contributions, annual bonus targets, private medical cover, and equity all have real value. Calculate what the total package is worth, not just the base. If the employer offers a strong bonus structure, a lower base anchor may be appropriate. If benefits are thin, your base anchor should compensate.
Fisher and Ury's foundational negotiation framework, Getting to Yes, makes a point that most people miss: the strongest anchor is not the highest number, it is the most credibly supported number. Citing a source shifts the conversation from a contest of wills to a discussion of data.
The practical application is simple. When you name your number, attach a reference immediately. Not "I think I deserve" or "I was hoping for." Instead: "Based on LinkedIn Salary data for senior software engineers in London and current live roles on Reed, the midpoint for this level sits at X. Given my experience and your sector, I am targeting Y."
That construction does three things. It makes your number look like the conclusion of a process, not a wish. It signals preparation. And it makes it harder for the employer to counter with an unsupported lower figure, because you have raised the standard of evidence in the room.
Do not cite more than two sources. More than two sounds defensive. Two sounds thorough.
Take a mid-level software engineer with four years of experience, based in London, targeting a role at a Series B fintech. Here is how the calculation works in practice.
Data collected:
Weighted baseline: applying higher weight to live adverts and LinkedIn, the baseline lands at approximately £65,500.
Adjustments applied:
Anchor stated in the negotiation: "LinkedIn Salary data for software engineers in London puts the midpoint at £65,000. Current fintech roles on Reed show a range of £55,000 to £80,000. Given four years of experience and the sector premium for fintech, I am targeting a base of £70,000."
That is a defensible, specific, credibly sourced anchor. It is not the ceiling of what is possible. It is the floor of a well-reasoned position.